doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 1 Project: IEEE P Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [TG3-Coding-criteria] Date Submitted: [11 January 2001] Source: [James P. K. Gilb] Company [Mobilian] Address [11031 Via Frontera, Suite C, San Diego, CA 92127] Voice:[ ], FAX: [ ], Re: [] Abstract:[The criteria for evaluating PHY coding methods for the P draft standard.] Purpose:[Describe the evaluation criteria for the voting members.] Notice:This document has been prepared to assist the IEEE P It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release:The contributor acknowledges and accepts that this contribution becomes the property of IEEE and may be made publicly available by P

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 2 Background The candidate draft standard was approved in November with the coding method for the higher order modulations (16-QAM, 32-QAM and 64- QAM) left undefined Four coding schemes were part of the propsals in Tampa, one has dropped (Rios). TG3 will vote to select one of three methods in Monterey The methods will be evaluated against the relevant criteria from P /110r14

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 3 Criteria Outline 1.Unit Manufacturing Cost 2.Delay Spread Resistance 3.Delivered Data Throughput 4.Range 5.Power Consumption 6.Latency 7.IP issues

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 4 Unit Manufacturing Cost UMC (section 2.1) addresses the PAR's requirement for a low cost system. Enabling lower cost implementations will increase the usefulness of the standard. All-digital implementations are on the favorable cost-curve of Moore's law. What is relatively expensive in bits now will be much cheaper in 4 years.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 5 Unit Manufacturing Cost Criteria Definition: The incremental cost of all parts required to implement a complete TX/RX coding solution including all logic and memory. Values: The incremental cost in US $ of implementing the proposed coding scheme calculated assuming $0.20 USD/100 kgates Success Criteria: Additional cost is less than ???

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 6 Delay Spread Resistance The wireless environment is hostile to high-speed and wide-band signals. The short-range, home environment, however, is more benign than office, factory or vehicular environments. Some coding schemes may improve delay spread resistance, others may not and rely instead on an equalizer.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 7 Delay Spread Resistance Criteria Definition: The delay spread tolerance is the value of T_RMS for which a maximum FER of 1% is met for 95% of the channels generated using the channel model defined in The power level at the transmitter is set 14 dB above the level required for a 1% FER in an AWGN channel. At least 1000 channels should be generated. Note that the channel model in will generate channels with fading parameters, so at the receiver the signal level will vary from one channel realization to the next channel realization. Values: The value of T_RMS in ns Success Criteria: The value of T_RMS is greater than 25 ns

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 8 Delivered Data Throughput The requirement for the standard is a system which can deliver 20 Mb/s. There is also a desire to have an enhanced mode with greater than 40 Mb/s. The coding methods will decrease the delivered data throughput while hopefully decreasing the PER.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 9 Delivered Data Throughput Criteria Definition: Delivered data throughput is the rate at which the user’s data is passed through the system. The values presented here assume that a microwave oven or other channel impairment will not be in operation at the same time as the desired signals are transmitted. If there is an operating microwave oven in the Personal Operating Space (POS) of this device, it is assumed that the user has enough control of the POS environment to turn it off when desiring to transmit. Values: Throughput in Mb/s and % overhead for each of the higher order modulations calculated using P /354r2 Success Criteria: None defined, more is better

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 10 Range The PAR requires a system that has a 10 m range in an environment where the user has control of potential interferers. The range depends on the actual fading, both flat and frequency-selective, the TX power, the RX sensitivity and the presence of other interfering devices. The actual range will be different at every location will change over time.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 11 Range Criteria Definition: Based on the PAR, the proposed system shall be able to initiate a WPAN connection within a 10 meter radius 99.9% of the time. Values: For each of the higher-order modulations, report the receiver sensitivity assuming a receiver noise figure of 12 dB and ideal isotropic antenna (i.e. 0 dBi gain). Success Criteria: Undefined, lower sensitivity is better

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 12 Power Consumption Wireless PAN connectivity implies small, mobile devices. Battery life will be an important aspect to the consumer Actual battery life will depend on the use model, which has not been defined by TG3. The goal is for the coding to have a minimal impact on the peak DC power consumption.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 13 Power Consumption Criteria Definition: The total amount of DC power required by the proposed system to operate the encoder or decoder for each of the higher order modulation schemes. Values: The DC power requirement for the encoder/decoder circuitry each of the higher-order modulations assuming m W/(MHz*kgate) Success Criteria: Undefined, lower is better. Total peak power for complete radio must be less than 0.5 W

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 14 Latency Latency was not specifically called out in the criteria document, but it was implied in the throughput requirement. A long latency in decoding or encoding a packet will affect the TX/RX turnaround time and hence degrade the efficiency of the system. A 10  s TX/RX turnaround time has been proposed. Fast decisions need to be done in hardware since software relies on interrupts which are relatively slow.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 15 Latency Criteria Definition: TX or encoding latency is the time from when the first bit to be encoded is presented to the encoder until the its representation is available at the output of the encoder. RX or decoding latency is the time from when the first demodulated symbol is present at the decoder until the first valid decoded bit is available at the output of the decoder. Values: TX and RX latency times in  s Success Criteria: Undetermined, less is better

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 16 IP Issues IP issues are not explicitly called out in the criteria document, but the IEEE has a very specific policy on it for standards The preference for a standard is for it to be unencumbered by essential IP claims Patented technology is allowed if it: –provides an important technical benefit –Is available in a non-discriminatory license for fair and reasonable terms.

doc.: IEEE /028r1 Submission 13 January 2001 James P. K. Gilb, MobilianSlide 17 IP Issues Criteria Definition: What known or potential IP claims are there for the proposed coding method. Values: A description of any potential IP or listing of the dates of published references to the method. Success Criteria: Bad: IP not available for fair and reasonable terms Good: IP owner is licensing or is willing to license Better: IP owner has accepted IEEE fair and reasonable Best: Method described in the open literature > 20 years ago